The core insight
Version 1 makes ethical work more profitable by combining funded rewards with scarce, valuable upside that unethical work cannot access. The exact condition is G < M(R+L): the deviation gain must be less than the scarcity multiplier times the total ethical benefit.
The Alignment Theorem
How do you make selfish agents behave ethically without forcing them?
Existing approaches each have known limits. Hand-coded rules are rigid and cannot adapt to novel situations. Learned policies can be gamed by the same optimization pressure that trained them. Centralized rulebooks concentrate control and create single points of failure. The Alignment Theorem proposes an economic mechanism instead: bind a declared community policy to verified facts and funded rewards, and let the incentive structure do the work. The protocol does not decide what is good; society does. It makes violations less profitable than compliance.
Three forces combine to make ethical work the most profitable choice.
The ethics signal: how the community scores behavior
Users publish formal ethical models in Tau Language. The network aggregates them into a single number: the Ethical-Eco Transaction Factor (EETF). This number, between 0 and 3, represents what the community currently considers ethical.
EETF is a community policy output, not an objective moral truth. A real deployment must specify who counts as a member, how votes are weighted, how to handle missing data, and who can change the rule.
Deflation creates scarcity
Version 1 assumes a shrinking spendable supply and an unbounded purchasing-power multiplier. Version 1.1 treats that path as an external economic scenario. Finite supply rules alone do not prove purchasing power diverges.
Pressure makes ethics profitable
Define R for direct reward exposure, L for scarcity upside protected by eligibility, and G for the complete bounded deviation gain. The historical negative “penalty” term is exactly the upside missed through exclusion. Moving that opportunity cost to the eligible side preserves the ordering and creates no punitive transfer.
The concrete V1 theorem and its V1.1 generalization.
Given:
- Authenticated eligibility: The policy root, network EETF, and candidate EETF tier are fixed for the decision.
- Enforceable exclusion: The eligible branch has funded direct reward coefficient R and exclusive scarcity-upside coefficient L. The excluded branch cannot obtain either entitlement. No tax or balance debit is applied.
- Strict finite margin: M(R + L) > G, and the agent exactly maximizes utility over the declared alternatives.
Then: Every exact maximizer selects the eligible branch. For integers and R+L > 0, the least strict multiplier is floor(G/(R+L))+1. If G=0, every positive M suffices.
V1.1 writes K(t)=R(t)+L(t) and places deviation gain, compliance cost, and optimizer error in B(t). If M(t)K(t)>B(t), every profit-maximizing strategy that comes close enough to the best payoff (within a small error margin) is eligible. Its relative-growth theorem covers time-varying opposing bounds.
In plain English: scarcity makes ethical work profitable only when the ethical reward grows faster than the gain from cheating.
Four obligations in the V1 exclusion argument.
Normalize exclusion correctly
The eligible branch earns MR (scarcity-amplified reward). The excluded branch earns G - ML (deviation gain minus the scarcity upside it cannot access). Adding ML to both sides shows the same comparison without any debit to the excluded branch.
Bind the exclusive entitlement
Authenticate EETF and scarcity, fund the reward, and prove that the excluded branch cannot claim L. Upside common to both branches cancels.
Require a strict margin
For V1 check M(t)[R(t)+L(t)] > G(t). V1.1 folds costs and optimizer error into B(t). Equality does not force an eligible choice.
Conclude within the model
The strict gap forces the eligible exact choice in V1 and the eligible near-optimal choice in V1.1. Both remain conditional on the declared alternatives and authenticated bounds. ∎
A feedback loop where ethical behavior raises rewards and rewards raise ethical behavior.
Behavior
EETF
Burns
Scarcity
The historical Virtuous Cycle Compounder (VCC) proposes three modules around this feedback loop:
- Dynamic Base Reward (DBR): The higher your ethical score, the more reward you earn for eligible work.
- Hyper-Compounding Rewards (HCR): The longer you stay ethical, the faster your rewards compound.
- Scarcity Buyback (AEB): Ethical participation triggers buybacks of the settlement asset, reducing its circulating supply and increasing its purchasing power.
If authenticated participation raises EETF, funded rewards are paid, buybacks reduce the circulating supply of a valued asset, and the protected upside remains exclusive, the loop can reinforce eligible behavior. Each arrow is a separate empirical or protocol premise.
The original experiments are exploratory. V1, V1.1, and V2 now have separate checked proofs.
Bounded legacy checks; no universal theorem claim.
Legacy scenarios remain exploratory. The current source candidate separately matched the V1, V1.1, and V2 packets; public-node deployment remains pending.
Zero failures across 691,000 randomized iterations
100% state coverage, no deadlocks detected
EETF tiers, opportunity-cost framing, the exact scarcity threshold, the eligible-choice guarantee, the unbounded-scarcity implication, and proof that upside shared by both branches cancels.
A local Tau build pinned to source commit 9b191af matched all 128 rows; only the all-true row is accepted. Facts remain externally supplied.
When the scarcity-amplified ethical reward strictly exceeds every modeled gain from cheating,
every near-optimal profit strategy favors the ethical action within the declared model.